The introduction of millions of electric vehicles (EVs) to the power grid will create a transformative moment for America’s decarbonization efforts. However, this also brings a significant challenge. Scientists and engineers are looking for the best way to ensure that vehicles can be charged smartly, efficiently, cheaply and cleanly through a grid that cannot accommodate them all at once or all the time.
Researchers at the US Department of Energy’s Argonne National Laboratory and graduate students at the University of Chicago are collaborating on an exciting new project to address that challenge. This project will use a particular combination of computational rewards and punishments—a technique called reinforcement learning—to train an algorithm to help schedule and manage the charging of a diverse set of electric vehicles.
The first group of vehicles the team studied were those charged by Argonne employees at the laboratory’s Smart Energy Plaza, which offers AC regular chargers and DC fast chargers. Since employees don’t usually need their cars during the work day, there is some flexibility in when each car can be charged.
“There’s a certain total amount of power that can be allocated, and different people have different needs for when they need to power their cars at the end of the day,” said Argonne principal electrical engineer Jason Harper. “Being able to train a model to work within the time constraints of a particular employee while knowing the peak demands on the grid will allow us to provide efficient, low-cost billing.”
“When you have many EVs charging at the same time, they create a peak demand at the power station. This introduces additional charges, which we are trying to avoid,” added Salman Yousaf. Yousaf is a graduate student in applied data science at the University of Chicago who worked on the project with three other students.
The reinforcement learning algorithm works by incorporating feedback from positive outcomes, such as an EV with the desired amount of compensation at the designated departure time. It also includes negative results, such as taking the power to a certain peak threshold. Based on this data, the charging scheduling algorithm can make more intelligent decisions about which vehicles to charge when.
“Smart charge scheduling is really an optimization problem,” Harper said. “In real time, the charging station often has to make tradeoffs to ensure that each car is charged as efficiently as possible.”
Although Argonne’s charging stations are the first location where project researchers are conducting reinforcement learning, there is potential to expand beyond the laboratory gates. “There’s a lot of flexibility when it comes to charging at home, where overnight charging will allow some ability to move around how the charging is distributed,” Yousaf said.
“Really smart billing is really taking into account all the actors in the ecosystem,” Harper added. “That means the utility, the charging station owner and the EV driver or home owner. We want to meet everyone’s needs while still being mindful of the restrictions everyone faces.”
Future work with the model will include a simulation of a larger charging network that will initially be based on data collected from Argonne’s chargers.
Harper and his colleagues have also developed a mobile app called EVrest that allows users of networked charging stations (in this case, former Argonne employees) to reserve stations and participate in smart charge scheduling. The EVrest platform collects data on charging behavior and will use that data to train future AI models to help with smart charge management and vehicle grid integration.
Provided by Argonne National Laboratory
Citation: Researchers improve electric vehicle charging management through machine learning (2023, July 24) retrieved 24 July 2023 from https://techxplore.com/news/2023-07-electric-vehicle-machine.html
This document is subject to copyright. Except for any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. Content is provided for informational purposes only.